ENVI EX Tutorial

Itt Visual Information Solutions

Informations

• Publié par: Mefer
• Nombre de lectures: 192
• Langue: English

Extrait

ENVI EX Tutorial: Image
Difference Change
Detection
I m a g e D i f f e r e n c e C h a n g e D e t e c t i o n W o r k f l o w 2
F i l e s U s e d i n t h i s T u t o r i a l 4
S e l e c t i n g F i l e s f o r I m a g e D i f f e r e n c e C h a n g e D e t e c t i o n 4
I m a g e D i f f e r e n c e C h a n g e D e t e c t i o n 4
C h a n g e T h r e s h o l d i n g 6
C l e a n i n g U p I m a g e D i f f e r e n c e C h a n g e D e t e c t i o n R e s u l t s 8
E x p o r t i n g I m a g e D i f f e r e n c e C h a n g e D e t e c t i o n R e s u l t s 9
1E N V I E X T u t o r i a l : I m a g e D i f f e r e n ce C h a n g e D e t e ct i o n
Image Difference Change Detection Workflow
In this tutorial, you will use the Image Difference Change Detection workflow to compare two images
of an area over Indonesia that was impacted by the December 26, 2004 tsunami. The first image is a
before image, taken in April, 2004. The second image was taken in January, 2005.
The first image shown below is from t s u n a m i _ b e f o r e . d a t, the second image shows t s u n a m i _
a f t e r . d a t in a Portal. You can see in the Portal that there are substantial differences between the
two images when you adjust the Transparency sliders on the toolbar. Specifially, many vegetation areas
were washed out by the tsunami.
2E N V I E X T u t o r i a l : I m a g e D i f f e r e n ce C h a n g e D e t e ct i o n
References
Image Difference Change Detection:
Normalized Difference Vegetation Index (NDVI): Jensen, J. R., 1986. I n t r o d u c t o r y D i g i t a l
I m a g e P r o c e s s i n g, Prentice-Hall, New Jersey, p. 379.
Normalized Difference Water Index (NDWI): McFeeters, S.K., 1996. The use of normalized
difference water index (NDWI) in the delineation of open water features, I n t e r n a t i o n a l J o u r n a l o f
R e m o t e S e n s i n g, 17(7):1425–1432.
Normalized Difference Built-up Index (NDBI): Zha, Y., J. Gao, and S. Ni, 2003. Use of
normalized difference built-up index in automatically mapping urban areas from TM imagery,
I n t e r n a t i o n a l J o u r n a l o f R e m o t e S e n s i n g, 24(3):583–594.
Burn Index: Burn Index uses an opposite Normalized Burn Ratio (NBR), which is -NBR. The
NBR reference is Key, C.H.; Z. Zhu; D. Ohlen; S. Howard; R. McKinley; and N. Benson, 2002.
The normalized burn ratio and relationships to burn severity: ecology, remote sensing and
implementation. In J.D. Greer, ed. Rapid Delivery of Remote Sensing Products. Proceedings of
the Ninth Forest Service Remote Sensing Applications Conference, San Diego, CA 8-12 April,
2002. American Society for Photogrammetry and Remote Sensing, Bethesda, MD.
Auto-thresholding:
Otsu's: Otsu, N., 1979. A threshold selection method from gray-level histograms. IEEE Trans.
Systems Man Cybernet. 9, 62–66.
3E N V I E X T u t o r i a l : I m a g e D i f f e r e n ce C h a n g e D e t e ct i o n
Tsai's: Tsai, W., Moment-preserving thresholding. Comput. Vision Graphics Image Process. Vol.
29, pp. 377–393, 1985.
Kapur's: Kapur, J., Sahoo, P., Wong, A., A new method for graylevel picture thresholding using
the entropy of the histogram. Comput. Vision Graphics Image Process. Vol. 29 (3), 273–285.
Kittler's: Kittler, J., Illingworth, J., Minimum error thresholding, Pattern Recogn. Vol. 19, pp. 41–
47, 1986.
Files Used in this Tutorial
ENVI Resource DVD: D a t a \ c h a n g e _ d e t e c t i o n
F i l e D e s c r i p t i o n
t s u n a m i _ b e f o r e . d a t QuickBird image over Indonesia, April, 2004
t s u n a m i _ b e f o r e . h d r Header file for above
t s u n a m i _ a f t e r . d a t QuickBird image over Indonesia, January, 2005
t s u n a m i _ a f t e r . h d r Header file for above
Selecting Files for Image Difference Change Detection
In the File Selection panel, you choose the two images to include in image difference change detection.
1. Start ENVI EX.
2. In the Toolbox, double-click I m a g e D i f f e r e n c e. The Select File panel appears.
3. Click B r o w s e next to the T i m e 1 F i l e field. The Select Time 1 Input File dialog appears.
4. Click O p e n F i l e. The Open dialog appears.
5. Navigate to D a t a \ c h a n g e _ d e t e c t i o n , select t s u n a m i _ b e f o r e . d a t, and click O p e n.
6. Click O K.
7. Click B r o w s e next to the T i m e 2 F i l e field. The Select Time 2 Input File dialog appears.
8. Click O p e n F i l e. The Open dialog appears.
9. Navigate to D a t a \ c h a n g e _ d e t e c t i o n , select t s u n a m i _ a f t e r . d a t, and click O p e n.
10. Click N e x t. The Image Difference panel displays.
Image Difference Change Detection
In the Image Difference panel, set the parameters to use for the difference analysis. In this step, you
perform image difference analysis based on a band or feature index. Feature index provides options to
detect changes of a specific feature, such as vegetation, water, built-up areas, or fire burn areas. For the
QuickBird data used in this exercise, Vegetation Inex (NDVI) and Water Index (NDWI) are available.
Built-up Index and Burn Index are available only if if an image has a shortwave infrared band, such as
Landsat data,
4E N V I E X T u t o r i a l : I m a g e D i f f e r e n ce C h a n g e D e t e ct i o n
1. In the D i f f e r e n c e B a n d tab, I n p u t B a n d and B a n d 1 were selected by default.
2. In the toolbar G o T o field, enter 7 4 6 3 1 9 . 4 9 9 , 5 8 5 3 0 3 . 4 7 1 and press the E n t e r key on the
keyboard. The Image window centers over the area.
3. Enable the P r e v i e w check box. A Preview Portal appears. In the Preview Portal, areas that
decreased in the data value of the selected band appear as red, and areas that increased appear as
blue.
4. With the Preview Portal still open, enable F e a t u r e I n d e x as the difference band and keep
V e g e t a t i o n I n d e x ( N D V I ) as the selected feature index.
5E N V I E X T u t o r i a l : I m a g e D i f f e r e n ce C h a n g e D e t e ct i o n
5. Disable the P r e v i e w check box, then click N e x t. The difference analysis begins.
6. When image difference processing is complete, the difference image appears in the Image
window and the Thresholding or Export panel appears.
7. Select A p p l y T h r e s h o l d i n g. This option allows you to set parameters that help the algorithm
determine which areas have big change. When you select this option, you can export multiple
outputs at the end of the workflow. (If you select E x p o r t I m a g e D i f f e r e n c e O n l y, you will not be
able to select additional processing parameters, and you can only export the difference image.)
8. Click N e x t. The Change Thresholding panel appears.
Change Thresholding
In the Change Thresholding step, specify change you want to show between the two images. You can
use pre-set auto-thresholding techniques, and you can manually adjust thresholding.
1. In the A u t o - T h r e s h o l d i n g tab, select I n c r e a s e a n d D e c r e a s e. This option shows areas of
increase (in blue) and decrease (in red). (If you are only interested in areas of vegetation
decreased by the tsunami, select D e c r e a s e O n l y.)
2. Enable the P r e v i e w check box. A Preview Portal opens.
3. In the S e l e c t A u t o - T h r e s h o l d i n g M e t h o d drop-down list, try selecting each option, one at a time,
then examine the result in the Preview Portal. The auto-thresholding choices are:
6E N V I E X T u t o r i a l : I m a g e D i f f e r e n ce C h a n g e D e t e ct i o n
l O t s u ' s : A histogram shape-based method. It is based on discriminate analysis and uses the
zeroth- and the first-order cumulative moments of the histogram for calculating the value of
the thresholding level.
l T s a i ' s : A moment-based method. It determines the threshold so that the first three moments
of the input image are preserved in the output image.
l K a p u r ' s : An entropy-based method. It considers the thresholding image as two classes of
events, with each class characterized by a Probability Density Function (PDF). The method
then maximizes the sum of the entropy of the two PDFs to converge on a single threshold
value.
l K i t t l e r ' s : A histogram shape-based method. It works on approximating the histogram as a
bimodal Gaussian distribution and finds a cutoff point. The cost function is based on the
Bayes classification rule.
The References at the beginning of this tutorial provide additional information about the auto-
thresholding methods.
4. In this exercise, we will use the default O t s u ' s thresholding method. Below is an example of the
Preview Portal with the O t s u ' s method selected.
5. You can also experiment with manually adjusting the threshold settings. To do this, select the
M a n u a l tab.
6. Use the slider bars to adjust the I n c r e a s e T h r e s h o l d and D e c r e a s e T h r e s h o l d settings, then
view the changes in the Preview Portal.
7E N V I E X T u t o r i a l : I m a g e D i f f e r e n ce C h a n g e D e t e ct i o n
7. When you are done experimenting with manual adjustments, click the R e s e t buttons to return
to the default settings.
8. Click N e x t. When you click N e x t, the difference image will be classified into B i g I n c r e a s e, B i g
D e c r e a s e and O t h e r, based on the th